JPH0744502B2 - CM voice level automatic control system - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Industrial field of use] The present invention relates to a CM for TV broadcasting, radio broadcasting, etc., and an automatic audio level control system for a recording studio, etc.

[Conventional technology]

It can be said that it is extremely difficult to keep the sound of all CMs audible and level constant when sending out CMs from CM banks.

Due to the nature of CM materials, sound effects, BG, narration, etc.
All sound sources are combined. In addition, there are many types of modes, such as mono, stereo, and two languages, and the level of the CM material that is delivered to the broadcasting station from the agency varies greatly. In recent commercials, with the shift to stereo, sound quality is becoming more important.

Under these circumstances, there are the following two methods that are generally adopted to keep the voice level of CM constant. That is, the first one is a method to keep the audio level constant by using hearing with the VU meter when recording the CM material, and the second is a constant audio level using the AGC or limiter compressor. How to keep.

[Problems to be Solved by the Invention]

However, since the first method relies on the VU meter to equalize the sound level, the level on the auditory sense does not match. For this reason, an operator who has undergone specialized training will work in combination with hearing, but there is a limit to that, and at the same time, there are individual differences.

For example, in the case of A-CM, B-CM, and C-CM as shown in FIG. 8, the operator is surprised to lower the voice level because the head part of A-CM has an impact. On the contrary, the whispering part in the latter half tends to raise the level. Also, as in the B-CM, the head part is low and the CM in the second half is high, the level rises on the contrary, so the level is over in the latter half, and the recording is NG. Furthermore, in CMs with a low average level, such as C-CM, the operator will slowly increase the fader. By the way, the simple average level of each CM is -1.2VU for A-CM, -2.8VU, C for B-CM.
-CM is -3.6VU.

The second method is difficult to determine the energy density of the signal,
I still have a complaint that the level of hearing is not uniform. In addition, whether it is an analog type or a digital type, it is basically a momentary response, and it is easy to cause an unnecessarily high level of movement and deterioration in sound quality, and in principle, it cannot be said to be faithful to the source sound.

In view of such conventional problems, an object of the present invention is to provide a control method capable of automatically and uniformly controlling all CMs in terms of level and audibility without impairing the source sound of each CM material. And

[Means for Solving the Problems]

The CM audio level automatic control system according to the present invention stores the level of the actual number of seconds of the CM audio when recording the CM material, obtains the audio level correction value by calculating the average level and the auditory sense correction, and at the time of reproduction. The variable attenuator is operated based on the voice level correction value to correct the CM voice level.

[Action]

First, when recording the CM material 1, the level of the actual number of seconds of the CM voice is stored in the storage unit. Next, the arithmetic unit calculates the average level and the auditory sense correction to obtain the audio level correction value.
Then, when the CM is reproduced, the variable attenuator is operated based on the sound level correction value to correct the CM sound level. This makes it possible to control the sound level of each CM material to a constant level without damaging the source sound.

〔Example〕

An embodiment of the control system of the present invention will be described in detail below with reference to the drawings.

FIG. 1 shows the first embodiment of the present invention, which is a case where the CM material is adopted in the MM system for recording in a master mother VTR (hereinafter referred to as MM-VTR).

In FIG. 1, reference numeral 1 is a CM material to be broadcast, which includes a VTR, a film, a slide and an audio tape, which are set in respective machines and reproduced. Reference numeral 2 is an MM-VTR, which is for recording all the CM material 1 brought in from the sponsor. Three
Is a correction value calculation device for storing the level of the actual number of seconds of the CM sound at the time of recording, calculating the average level and the auditory sense correction, and obtaining the sound level correction value. 4 is an on-air mother group, which contains all the commercials used for broadcasting. 5 is a variable attenuator, 6 is a host computer,
When recording the voice level correction value and listening to check the recording status of the MM-VTR2, or recording on the on-air mother group 4, give an instruction to the variable attenuator 5 based on the correction value. The variable attenuator 5 is operated to correct the CM voice level.

When recording CM material 1 in MM-VTR2, use the reference signal for audio level adjustment. Then, in order to confirm the recording status, a full audition is performed. Therefore, the sound level correction value is obtained by the correction value calculation device 3 at the time of recording and recorded in the host computer 6, and correction is made based on the correction value at the time of trial listening. Therefore, the sound level of any CM material 1 can be auditioned at a constant volume. In addition, when recording to the mother group 4, it is possible to record at a constant volume.

Next, the correction value calculation device 3 will be described. As shown in FIG. 2, the correction value calculation device 3 converts the voice signal of the CM voice into
60Hz-600Hz reduction, 600Hz-1.6KHz middle range and 1.6KHz-
A filter unit 7,8,9 that divides into three high frequency bands of 16KHz, and A / D converters 10,11 and 12 that perform A / D conversion of the audio signal of each of the three divided bands, A storage unit 13, 14, 15 for storing each digital audio signal that has been D / D converted, and an arithmetic unit 16, 17, for obtaining an average value for each frequency band from each digital audio signal, and weighting for hearing correction It is provided with 18 and a total calculation unit 19 that calculates the sound level correction value by comprehensively calculating the calculation results. By the way, low range is 240Hz, mid range is 240Hz ~
6.4KHz, high frequency is set to 6.4KHz. And each A /
The D conversion units 10, 11, 12 and the storage units 13, 14, 15 are controlled to start and end via the control unit 20 by a control input from the outside, and the calculation units 16, 17, 18 are used to start the calculation. The control is performed, and further, the total calculation unit 19 is controlled to end the calculation.

When a voice signal of a CM voice is input at the time of audition or the like, it is passed through the filter units 7, 8 and 9 and divided into three for each frequency band to cut excess signals. Next, the three-divided audio signal is rectified and digitized by the A / D converters 10, 11 and 12, respectively. The sampling frequency at this time may be a frequency at which average value setting and audibility correction can be performed, but 4 Hz is sufficient.

That is, in ordinary conversation, as shown in FIG.
~ 10 syllables with alternating periods of no and no sound. And the voice energy of conversation is very low. Therefore, since the main part of the program is conversation, the commercial is broadcast 6 dB down compared to the main part of the program. The comment-only CM is the reason why we are raising the level.

The voice band for conversation can be considered to be 6 KHz as shown in FIGS. 4 (A) and 4 (B), which shows the long-term average spectrum of Japanese people.

On the other hand, there is BG as an important element that constitutes CM, and the spectrum of the light music and the orchestra is shown in FIGS. 5 (A) and 5 (B). From FIGS. 5 (A) and 5 (B), 10
It can be seen that a band of KHz or higher is required, but 4Hz is sufficient for the purpose of average value calculation and hearing correction.

The following table shows the average values of various parameters of various records.

As you can see, the average level is popular at -6.
It is 3 dB. In the case of CM, the BG goes down and narration is put on it, so the average level will rise further.

Since each A / D converter 10, 11, 12 samples the audio signal at a rate of 4 times per second and digitizes it, each storage unit 1
3, 14 and 15 store the level value at a rate of 4 times per second, respectively. At this time, the maximum number of CM seconds of voice stored is 3 minutes.

When the recording starts, the A / D converters 10, 11, 12 and the storage units 13, 14, 15 start operating, and when the recording ends, the recording end signal at that time causes the calculating units 16, 17, 18 to The calculation unit 19 starts the calculation operation. Then, the calculation units 16, 17, 18 reduce the low / medium /
The average value of the voice level is calculated for each high frequency band,
A weight is added for the auditory sense correction. The weighting coefficient setting at this time is arbitrary. When calculating the average value, silence without voice input is not calculated and is deleted from the average value calculation data. This is a commercial that intentionally silenced a part
This is because, in the case of a still image CM, since there is a lot of silence because the voice is synchronized with the screen, simply averaging the seconds decreases the average value.

The average value calculation and the auditory sense correction calculation are performed as follows.
FIG. 6 shows the relationship of normal sound pressure level (SPL). From now on, the maximum value is 120 dB, but the listening level when each person listens at home can be considered to be 60 to 70 dB from the following table.

The reason why this listening level is required is that the loudness of the sound is related to the strength of the sound, but is greatly affected by the kind of the sound (difference in frequency components). For example
When you hear the sound pressure level of 40Hz at 1000Hz and 100Hz, the sound at 1000Hz is 10 even though the sound pressure level is the same.
It sounds much louder than the 0 Hz sound. On the other hand, when the sound pressure level increases, the difference due to the frequency decreases. In this way, the loudness of sound is proportional to the frequency and intensity.

FIG. 7 shows the audibility correction curve of the sound level meter, and shows the basic form of three characteristics of A characteristic, B characteristic, and C characteristic.

The A characteristic is 60 dB or less, the B characteristic is 60 to 80 dB, and the C characteristic is 85 dBB or more.

As described above, since the listening level at home is 60 to 70 dB, the B characteristic is a curve for hearing correction. That is, according to the B characteristic, it is necessary to correct about 5 dB at 100 Hz. At this time, if the A characteristic is used, 100 Hz
Therefore, a correction of about 2 dB is required, and if the C characteristic is used,
No need to correct.

In general, therefore, when viewing television, the volume of the B characteristic is considered to be the majority, so if the weighting coefficient of the B characteristic is used, it should agree with the feeling of hearing. However,
Since the actual CM includes all frequencies, the average level of all frequencies will cause a difference from the sensation of hearing.

That is, in the case of CM, because it is composed of BG and narration,
It seems that the frequency components and the strength are in the entire band, so how to make the audible correction electrically becomes a problem.
However, since there is already a sound level meter, it can be applied. In other words, CM can be called a kind of noise.

It should be noted that the audibility correction curve and the actual commercial may not match in terms of audibility. For example, there are cases where the correction of the high frequency region is more audibly consistent than the correction of the low frequency region. Therefore,
In that case, it suffices to lower only the high frequency band by, for example, -3 dB. Therefore, it is desirable that the auditory sense correction value can be set arbitrarily.

That is, the arithmetic unit 3 first divides the voice into three parts, a low band, a middle band, and a high band, and calculates the average value of each. Next, each average value is compared with each target value. That is, the weighting is determined according to the change of the target value, and how much the correction should be made to obtain each target value. This is the auditory sense correction, which means the weighting.

Therefore, each weighting coefficient is arbitrarily set by the three computing units 16, 17, and 18 in accordance with the desired (desired) target value.

Then, the total arithmetic unit 19 is configured to operate the above-mentioned three arithmetic units 16, 17, 18
The calculated values from are averaged, and the calculated value is calculated as a voice level correction value.

The variable attenuator 5 receives the above-mentioned audio level correction value, and adjusts the audio level at the time of trial listening based on this value.
To correct. That is, the variable attenuator 5 receives the audio level correction value calculated by the above-mentioned general calculation unit 19, and at the time of audition, "the part that the sponsor wants to emphasize and the part that the user wants to suppress can be appropriately expressed. Moreover, it is possible to set the sound level to the viewer so that the discomfort of the commercial can be eliminated.

When the voice level correction value is obtained in this manner and the correction is performed based on this, the result is as shown in FIG. This FIG. 9 is a correction of each CM shown in FIG.
Is 2 dB lower, B-CM is 1 dB lower, C-CM is the same, and both are average levels of -3.5 dB.

Therefore, by performing the average level and the auditory sense correction in this way, it is possible to appropriately express the portion to be emphasized and the portion to be suppressed to a low level according to the intention of the sponsor, and to eliminate the discomfort of the CM from the listener. This method has the property of a CM bank (once filing, it is used repeatedly.
The data and materials are all managed by a computer), and it is possible to perform digital calculation with a microcomputer or the like.

Although the MM method has been described in the above embodiment, the present invention is not limited to this, and the same can be applied to the direct filing method as shown in FIG. 10, for example. Also, radio broadcasting, recording studios, etc. can be applied freely.

〔The invention's effect〕

According to the present invention, the average level of CM voice and the auditory sensation correction are calculated to obtain a voice level correction value, and the variable attenuator is operated based on the level correction value to correct the CM voice level. Even if each CM material is recorded without consideration,
You can automatically maintain a uniform level for all commercials, and
It does not spoil the source sound of CM material.

[Brief description of drawings]

FIG. 1 is an overall block diagram showing an embodiment of the present invention, FIG. 2 is a block diagram of a correction value computing device thereof, FIG. 3 is a voice waveform diagram, and FIGS. 4 (A) and 4 (B) are Japanese. Figures showing human long-term average spectrum, Figures 5 (A) and (B) showing BG spectrum, Figure 6 showing sound pressure of normal sound, Figure 7
FIG. 8 is a characteristic diagram of auditory sense correction, FIG. 8 is a diagram showing a sound level at the time of recording, FIG. 9 is a diagram showing a sound level after correction, and FIG. 10 is a block diagram showing another embodiment of the present invention. is there. 1 ... CM material, 3 ... correction value calculation device, 5 ... variable attenuator, 6 ... host computer.

Claims (1)

[Claims] 1. When recording a CM material, the level of the actual number of seconds of the CM voice is stored, the average level and the audible correction are calculated to obtain the voice level correction value, and the voice level correction value is based on the voice level correction value during reproduction. The automatic CM voice level control method is characterized by activating a variable attenuator to correct the CM voice level.